sayan1999
/ chatterbox_fp16_models.py
Last active
June 12, 2026 11:58
ChatterboxTurboTTS GPU FP16 Quantization Module
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| """ | |
| Thin subclasses of the Chatterbox TTS models with partial FP16 quantization. | |
| t3 + ve → FP16 (autoregressive transformer tolerates half precision) | |
| s3gen → FP32 (flow matching + HiFiGAN vocoder — kept full precision for audio quality) | |
| """ | |
| import torch | |
| from chatterbox.tts_turbo import ChatterboxTurboTTS as _Turbo |